2014/11/01 by Stefano Castruccio, Raphaël Huser, Castruccio, Stefano +4 · 5 citations
Decision Sciences · Economics, Econometrics and Finance · Engineering · Mathematics · #Advanced Statistical Process Monitoring #Algorithm #Artificial intelligence #Composite number #Computer science #Econometrics #Estimator #Fault Detection and Control Systems #Financial Risk and Volatility Modeling #Inference #Likelihood function #Machine learning #Mathematics #Maximum likelihood #Multivariate statistics #Quasi-maximum likelihood #Statistical Methods and Bayesian Inference #Statistical Methods and Inference #Statistical inference #Statistics #Truncation (statistics) #stat.CO
paper · pdf · doi:10.48550/arxiv.1411.0086
published in arXiv (Cornell University) (Cornell University) · in Journal of Computational and Graphical Statistics, 2016
openalex publication_date 2014/11/01 · arxiv created 2015/08/19 · arxiv updated 2015/08/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
In multivariate or spatial extremes, inference for max-stable processes observed at a large collection of points is a very challenging problem and current approaches typically rely on less expensive composite likelihoods constructed from small subsets of data. In this work, we explore the limits of modern state-of-the-art computational facilities to perform full likelihood inference and to efficiently evaluate high-order composite likelihoods. With extensive simulations, we assess the loss of information of composite likelihood estimators with respect to a full likelihood approach for some widely used multivariate or spatial extreme models, we discuss how to choose composite likelihood truncation to improve the efficiency, and we also provide recommendations for practitioners. This article has supplementary material online.